290 Mr. H. J. Carter on some West-India7i 



generally of the skeletal spicule is so far peculiar iu itself that 

 a practised eye can almost always recognize its Esperian cha- 

 racter. 



Size. — In measuring tliese spicules, again, great care should 

 be taken ; for here as well as elsewhere it should never be 

 forgotten that things must be small before they are great ; 

 hence both skeletal and flesh-spicules of all sizes below the 

 average largest may be present in the specimen ; hence the 

 necessity of finding out the average : thus, the so-called ^' ten- 

 sion-spicuia," viz. figs. 1(3 and ^6 in Dr. Bowerbank's illus- 

 trations of Efijieria [llhapliiothsma) florea cind Uiigua resj^ec- 

 tively, appear to be only small forms of the skeletal spicules 

 (tigs. 15 and 2), which, as the dermal layer becomes part of the 

 internal structure in the course of growth, become enlarged 

 to the size of skeletal ones. 



Rosettes. — The well-known "rosettes" Avliich characterize 

 the s]uculation of Esperia, viz. the globular develo])ment of a 

 multitude of f^^equianchorates (instead of a single one in a cell, 

 as with the bihamates and tricurvates, &c.) , wliich radiate from 

 a common centre with their small ends inwards, is not always 

 confined to the /»equianchorate flesh-spicules ; for the same 

 kind of development may occur in Desmacidon tituhans, 8dt,, 

 where the anchorates are equally developed at each end, as 

 seen in Schmidt's mounted type specimen of this sponge in 

 the British Museum (PI. Xll. fig. 24, _$',/<). No one, how- 

 ever, has described and illustrated the development of the 

 " rosette " — that is, the /??equiancliorate in ]:)lurality in its cell ; 

 although singly it has been done by Schmidt and myself in- 

 dependently (Nord-See Exped. 1872, " Zoologie," Taf. i. ; 

 and ' Annals,' 1874, vol. xiv. p. 100, pi. x.). 



Lastly, there is a characteristic dermal structure in Esjjeria 

 which for uniformity and beauty of its stellitication equals, if 

 not surpasses, any other of the kind. This consists of a stelli- 

 ferous lacework formed by intercrossing bundles of the 

 skeletal spicules (whose interstices when fresh are tyrapanized 

 by the dermal sarcode in which the pores are situated), 

 supported by a more or less rigid spiculo-iibrous structure 

 internally, that, especially when rigid, is equally characteristic 

 oiEsperia. Sometimes, however, the " lacework " structure 

 of the surface seems, from some cause or other, to become a 

 broken-down or confused layer of spicules, in which state the 

 tvi'o conditions may be seen to pass into each other in the same 

 specimens ; or the dermal layer together with the softer struc- 

 ture filling the interstices of the rigid skeletal fibre may be 

 washed away altogether, while the latter remains in a naked 

 condition (see Schmidt's representation of Esi>eria Contareniiy 



